Power calculation from library cell data

advanced_asic_chip_synthesis_-_bhatnagar.pdf

Hello!

I was looking the other day at a .lib file for some libraries used, trying to find out how power is calculated by power estimatiob tools out of the library data provided
for each cell.

So, for each cell the library provides the following:

internal power which has two components fall_power and rise_power and
leakage power. example:

cell (XXXXXX) {
area : 598.6896 ;
cell_footprint : blablabla ;
dont_touch : true ;
..
..
..
delay_tables here
..
..
internal_power () {
related_pin : "CLKIN" ;

fall_power (energy_template_2x2) {
index_1 ("....") ;
index_2 ("....") ;
values ("....") ; <--table
}
rise_power (energy_template_2x2) {
index_1 ("...") ;
index_2 ("...") ;
values ("....") ; <--table
}
}
}
cell_leakage_power : 223502.753119 ;


Ok, now cell leakage power is pretty straightforward. Leakage for specified PVT. But what about internal power? The only power component that I know dependent on rise and fall times is Short-Circuit power or crowbar current power for CMOS circuits.

Does this mean that within a cell only short-circuit power and leakage are calculated and then dynamic(switching power) is calculated out of the switching activity per interconnect and capacitances out of wire load models?

To which power components do these info in the library correspond?

Thanx!!

how is leakage calculation done from a library

I think you are thinking on the right lines. I would recommend you to read Himanshu Bhatnagar book for more details. The book is

Advanced ASIC Chip Synthesis”, Second Edition, Kluwer Academic Publishers

**broken link removed**

power calculation cell

thanx for your reply!

I have this ebook and started reading it but got dissapointed with the way it is written. Moreover some things it provides are outdated, so i quit...

Instead I started reading synopsys manuals to get a good idea of the tools and I think that was a wise decision to make..

Regarding power, my thoughts are based on what I have read in various books and synopsys manuals as well.

I just am not sure if rising_power and falling_power corresponf to data for short-circuit or crowbar currents power dissipation.

So i could read some more opinions on that by more people...

Thanks again!